Functional Analyses of Distinct Oxygen Concentrations and Light Niches of Prochlorococcus marinus
Mireille Savoie, Julie Nadeau, Laurel Genge, Sylwia Śliwińska-Wilczewska, Maximilian Berthold, Naaman Omar, and Douglas A. Campbell
2024-01-18
1 Department of Biology, Mount Allison University, 53 York St., Sackville NB, Canada, E4L 1C9.
Abstract
Introduction
Materials and methods
Culturing
%%%%%% Taken from my thesis word for word, need to reword and shorten %%%%
Three xenic P. marinus cultures, obtained from Bigelow Labs, NCMA Maine, US: MED4 (CCMP1986) from a High Light-adapted (HLI) clade, SS120 (CCMP1375) from a Low Light-adapted (LLIII) clade and MIT9313 (CCMP2773) also from a Low Light-adapted (LLIV) clade. These cultures were then maintained in two separate incubators. The temperature for both incubators was set to 22°C and a light/dark cycle of 12 h. The PAR level of the incubator was chosen to reflect the light level of the natural niche of the ecotype during culturation. The PAR level of the incubator containing the HLI clade, MED4, was set at 160 µmol photons m-2 s-1, whereas the incubator containing the LLIII and LLIV clades, SS120 and MIT9313 respectively, was set at 30 µmol photons m-2 s-1. To ensure cultures remained in exponential growth phase, all strains were transferred weekly in Pro99 media prepared according to [1] in autoclaved artificial seawater. Artificial seawater was prepared according to the National Center for Marine Algae and Microbiota (NCMA) protocol by combining salt solution I and salt solution II using the enriched artificial seawater (ESAW) recipe.
Experimental design
%%%%%% Taken from my thesis word for word, need to reword and shorten %%%%
For each of three ecotypes, we imposed [O2] (three levels: 2.5 µM, 25µM, 250 µM), photoperiod (four levels: 4 h, 8 h, 12 h, 16 h), spectral waveband (four bands: full spectrum white LED, 660 nm, 530 nm, 450 nm), and light level (three levels: 30, 90, 180 µmol photons m-2 s-1) treatments in a factorial design. Each factor is explained below. The full crossing of all factors would yield 3 x 4 x 4 x 3 = 144 treatments per ecotype (432 total), but due to time constraints and total absence of growth of some ecotypes under some conditions, not all treatments were carried out. In total, we completed 291 treatments.
All growth experiments were conducted at 22°C.
Experimental light conditions
%%%%%% Taken from my thesis word for word, need to reword and shorten %%%%
Three growth Photosynthetically Active Radiation (PAR) levels (180, 90, and 30 µmol photons m-2 s-1) and four spectral wavebands (white LED full spectrum, 660 nm, 530 nm, and 450 nm) were chosen to simulate light levels and spectral color spanning the vertical ocean water column, from near-surface to the lower euphotic zone depths (Figure ??). For simplicity, actinic light used for growth under specific wavebands will be referred by the respective spectral color; white LED for LED full spectrum, red for 660 nm, green for 530 nm and blue for 450 nm. Four different photoperiods were chosen to simulate various diel cycles characteristic of current and hypothetical future niches of P. marinus. A photoperiod of 16 h was chosen to represent temperate (45°N) summer at the ocean surface, 12 h for equatorial (0°N) ocean surface or temperate (45°N) spring and fall ocean surface or temperate (45°N) summer at deep ocean depths, 8 h for temperate (45°N) winter at the surface or at temperate (45°N) spring and fall at depth and equatorial (0°N) deep ocean depths and 4 h for temperate (45°N) winter or deep ocean depths during temperate (45°N) spring and fall.
Experimental oxygen conditions
%%%%%% Taken from my thesis word for word, need to reword and shorten %%%%
Three target dissolved oxygen concentrations [O2] were delivered to tubes of the Multicultivator by mixing varying ratios of air and Nitrogen (N2) gases while delivering 0.05% of Carbon Dioxide (CO2) gas through a 0.2 μm sterile microfilter via a G400 gas mixing system. To confirm and monitor the [O2], 4 FireSting optodes (PyroScience, Germany) were inserted into select tubes of each modified [O2] run for real-time measurements. A compensation temperature probe was placed in the aquarium of the bioreactor to correct [O2] for temperature fluctuations. In addition, the software corrected [O2] based on the salinity of the media (32 ppt). For the low O2 environment experiments, 0.5 to 1.0 µM of O2 was delivered to each Multicultivator tube by sparging with a gas mixture containing 99.95% N2 and 0.05% CO2 to purge dissolved O2 out of the culture. The intermediate O2 environment experiments were sparged to deliver 10 to 25 µM of O2 using a gas mixture containing 98.95% N2, 0.05% CO2 and 1% O2. The high O2 environment experiments were sparged with lab air (78% N2, 21% O2, 1% Ar and 0.05% CO2) to deliver 230 µM of O2. While the flow rate of the gas mixture was controlled, variations in bubbling speed affected the [O2] delivered to each tube; therefore, a range of [O2] was defined for each experimental O2 level; 0.5 µM - 5 µM for low, 5 µM - 50 µM for intermediate and 200 µM - 280 µM for high O2 experiments. To simplify the representation of experimental [O2] conditions in graphs and discussions, the approximate median [O2] of each experimental range: 2.5 µM, 25 µM, and 250 µM, will be used for low, intermediate, and high conditions, respectively. Figure ?? shows the data capture software of the FireSting Oxygen Logger. Dissolved [O2] were measured every 5 minutes over the duration of the Multicultivator run and recorded in a text file for later processing in R-Studio (Figure ??).
Bioreactors
%%%%%% Taken from my thesis word for word, need to reword and shorten %%%%
Growth experiments under different spectral wavebands were performed using a PSI MCMIX-OD Multicultivator (Figure 1) or a PSI MC1000-OD Multicultivator for white LED experiments. Each Multicultivator has the capacity to individually control 8 tubes with specific PAR levels and photoperiods and the MCMIX-OD has options for individually controlled spectral wavebands. 10 mL of exponential growth culture was added to 70 mL of Pro99 media and all 8 tubes were situated in a common temperature-controlled water bath to ensure the temperature remained constant at 22°C over the duration of the experiment. Real time absorbance measurements of Optical Density (OD) 680 nm (a proxy for cell suspension density, cell scatter and cell chlorophyll content) and OD 720 nm (a proxy for cell suspension density and cell scatter) were recorded every 5 minutes for at least 8 to 14 days depending on the duration of the lag phase, if any. Figure ?? is a typical readout from the MCMIX-OD Multicultivator software. Real time OD measurements eliminate intrusive subsampling of sterile cultures and provide high resolution chlorophyll and cell scatter proxies over the duration of the experiment. All data from the Multicultivator were saved as a comma separated values file and processed in R-Studio for calculations of growth rate estimates and graphical plotting.
Figure 1: PSI MCMIX-OD Multicultivator. The image illustrates the capability to set different spectral wavebands and light levels for individual culture tubes. Tubes 4 and 7 have oxygen optodes inserted for real-time dissolved oxygen concentration measurements. Real time Optical Density (OD) measurements eliminate intrusive subsampling of sterile cultures.
Data management and analysis
%%%%%% Taken from my thesis word for word, need to reword and shorten %%%%
Data from the Multicultivators were imported into R-Studio for data
management, growth rate calculations, statistics, and generation of
figures with the ggplot2 package [2]. The chlorophyll proxy
optical density (OD680 - OD720) or ΔOD was used to determine the growth
rate for each condition. We used a rolling mean from the RStudio zoo
package [3] to calculate the average ΔOD data over a 1-hour window.
This was done to prevent extraneous data points from affecting the
growth rate estimates. A Levenberg-Marquardt algorithm
[4] modification of the non-linear
least squares fit equation using the R package minpack.lm
[5] was used to calculate growth rate (µ) using the logistic
equation (1):
where ΔODmax is maximum ΔOD, ΔODmin is minimum ΔOD, t is time duration over the growth trajectory. Figure ?? is an example of a chlorophyll proxy growth estimates fitted from the high resolution ΔOD measurements for each tube in a Multicultivator. The residuals of the logistic growth curve fit are shown and the growth spectral waveband is plotted and illustrates the imposed PAR (µmol photons m-2 s-1) and photoperiod (h). Although the chlorophyll proxy (ΔOD) growth rate was used in this study, we also determined the cell scatter proxy (OD720) growth rate. The correlation between the chlorophyll proxy growth rate and the cell scatter proxy growth rate of P. marinus under all conditions examined in this study was shown in Figure ?? and generally showed balance growth. The high cell scatter growth rate of cultures that exhibited no chlorophyll proxy growth may be the logistic model over-estimating the fits as the amplitude of the OD720 signal maybe very small.
Generalized additive model
%%%%%% Taken from my thesis word for word, need to reword and shorten %%%%
A Generalized Additive Model (GAM) was applied to examine the
relationship between the chlorophyll proxy (ΔOD) growth rate across the
blue spectral waveband, photoperiod and PAR levels for each P.marinus
ecotype in this study. The gam function from the R package mgcv
[6] was used to model the growth rate with smoothing terms to
indicate the 90, 50 and 10% quantiles. Only data below a standard error
tolerance of 30% of the fit was used in the model. Because of time
limitations, we were unable to conduct sufficient growth response
experiments for all the other spectral wavebands, except for the blue,
to fulfill the input requirements for the GAM. Therefore, our priority
was on studying the effect of blue light on growth trends, considering
that blue light is the most ecologically relevant spectral waveband for
deep ocean niches.
Results
Figure 2: fmole target protein per ug total protein for Prochlorococcus marinus MED4 (High Light (HLI) near surface clade). Growth Photosynthetically Active Radiation (PAR) (µmol photons m-2 s-1) and spectral wavelength are in rows; 2 levels of imposed growth dissolved O2 concentrations (µM) are in columns. Numbers over each bar are fmole/ug
Figure 3: fmole target protein per ug total protein for Prochlorococcus marinus MIT9313 (Low Light (LLIV) deep ocean clade). Growth Photosynthetically Active Radiation (PAR) (µmol photons m-2 s-1) and spectral wavelength are in rows; 2 levels of imposed growth dissolved O2 concentrations (µM) are in columns. Numbers over each bar are fmole/ug
Supplemental
Figure 4: Chlorophyll proxy growth rate (d-1) for Prochlorococcus marinus MED4 (High Light (HLI) near surface clade) vs. photoperiod (h). 3 levels of growth Photosynthetically Active Radiation (PAR) (µmol photons m-2 s-1) are in columns; 3 levels of imposed growth dissolved O2 concentrations (µM) are in rows. Colors represent the actinic spectral waveband (nm). Large circles show mean or single determinations of growth rate from logistic curve fits (ex. Figure ??); small circles show values for replicate determinations, if any: replicates often fall with larger circles
Figure 5: Chlorophyll proxy growth rate (d-1) for Prochlorococcus marinus SS120 (Low Light (LLIII) deep ocean clade) vs. photoperiod (h). 3 levels of growth Photosynthetically Active Radiation (PAR) (µmol photons m-2 s-1) are in columns; 3 levels of imposed growth dissolved O2 concentrations (µM) are in rows. Colors represent the actinic spectral waveband (nm). Large circles show mean or single determinations of growth rate from logistic curve fits (ex. Figure ??); small circles show values for replicate determinations, if any: replicates often fall with larger circles
Figure 6: Chlorophyll proxy growth rate (d-1) for Prochlorococcus marinus MIT9313 (Low Light (LLIV) deep ocean clade) vs. photoperiod (h). 3 levels of growth Photosynthetically Active Radiation (PAR) (µmol photons m-2 s-1) are in columns; 3 levels of imposed growth dissolved O2 concentrations (µM) are in rows. Colors represent the actinic spectral waveband (nm). Large circles show mean or single determinations of growth rate from logistic curve fits (ex. Figure ??); small circles show values for replicate determinations, if any: replicates often fall with larger circles